Palladium-catalyzed enantioselective oxidation of chiral secondary alcohols: Access to both enantiomeric series

Article abstract of DOI:10.1002/anie.200801865

(Chemical Equation Presented) Rapid resolution: A new catalyst system for the oxidative kinetic resolution of secondary alcohols leads to dramatic rate increases. This system allows the use of a diamine to provide access to either enantiomer of a range of alcohols with good selectivity factors (see scheme). This method has been applied to the formal total synthesis of (-)-amurensinine.

Full text of DOI:10.1002/anie.200801865

Angewandte
Chemie
DOI: 10.1002/anie.200801865
Kinetic Resolution
Palladium-Catalyzed Enantioselective Oxidation of Chiral Secondary
Alcohols: Access to Both Enantiomeric Series**
David C. Ebner, Raissa M. Trend, CØdric Genet, Matthew J. McGrath, Peter OꢀBrien,* and
Brian M. Stoltz*
Enantioenriched alcohols are ubiquitous in the structures and
syntheses of natural products and pharmaceuticals. Catalytic,
asymmetric alcohol oxidation can be a useful method to
access these molecules.[1] Previously, we reported the devel-
opment of an aerobic kinetic resolution of alcohols by
catalytic [Pd(nbd)Cl2] (1, nbd = norbornadiene) and the
naturally occurring alkaloid (À)-sparteine ((À)-2) in the
presence of molecular oxygen.[2–6] Although this system
successfully resolves a wide range of secondary alcohols to
high enantiomeric excess, the rates of oxidation for certain
substrates are prohibitively slow. Furthermore, the use of 2 as
a ligand, which is only commercially available as the
(À)-antipode, restricts access to alcohols in one enantiomeric
series.[7] Herein, we disclose the development of a catalyst
based on an understanding of the reaction mechanism that
effects dramatic rate increases, thereby permitting resolution
of a broader range of substrates. This more active catalyst
allows the use of an alternative chiral diamine ligand in the
resolution, making either enantiomer of the secondary
alcohols easily obtainable. The utility of the system is
demonstrated in the formal total synthesis of naturally
occurring (À)-amurensinine ((À)-3).
teine ligand induces a significant distortion of the square
planar geometry in many palladium complexes (Scheme 1).[8]
Specifically, for the solid-state structure of [Pd(sp)Cl2] (4), the
sum of the six angles around the metal center is 705.998,[8]
compared to 7208 for an ideal square planar geometry.[10] The
Scheme 1. Model of the alcohol oxidation with [Pd(sp)X2].
Our initial screens of various Pd sources revealed the
dichloride complexes to be superior to the acetate and
trifluoroacetate species.[2] X-ray crystallographic analyses of a
series of crystalline palladium(II) complexes[8] and computa-
tional studies of mechanistic pathways[9] led to a better
understanding of the role of the halide counterion in the
resolution. The sterically crowded, C1 symmetric (À)-spar-
majority of this distortion is because of the deflection of X2
away from the projecting piperidine ring of (À)-2. For
dichloride complex 4, X2 is 9.98 out of the plane. This
deformation is even more pronounced in the structure of a
palladium alkoxide that mimics a proposed alcohol oxidation
intermediate (7, R1 = Ph, R2 = CF3, X2 = Cl, sum of six
palladium–ligand angles: 701.588, X2 deflection: 15.48).[8]
The less active [Pd(sp)(OAc)2] catalyst (5) for the kinetic
resolution[11] has a smaller deviation from the ideal square
planar geometry (sum of six palladium–ligand angles: 711.408,
X2 deflection: 5.38).[12] This (À)-sparteine induced distortion
of X2 results in a geometry that is more like the transition
state (8),[9] potentially lowering the energy barrier to
b-hydride elimination. Thus, we predicted that palladium
complexes with coordinated counterions that display a
greater X2 deflection would serve as more active oxidation
catalysts.
[*] C. Genet, Dr. M. J. McGrath, Prof. P. O’Brien
Department of Chemistry
University of York
Heslington, York, YO10 5DD (UK)
Fax: (+44)1904-43-2516
E-mail: paob1@york.ac.uk
Homepage: http://stoltz.caltech.edu
D. C. Ebner, Dr. R. M. Trend, Prof. B. M. Stoltz
The Arnold and Mabel Beckman Laboratories of Chemical Synthesis
Division of Chemistry and Chemical Engineering
California Institute of Technology
1200 E. California Blvd., MC 164-30, Pasadena, CA 91125 (USA)
Fax: (+1)626-564-9297
This hypothesis inspired us to investigate bromide as a
larger, but still coordinating counteranion. X-ray crystallo-
graphic analysis of a single crystal of [Pd(sp)Br2] (6)[13]
revealed a greater deviation of one of the bromides from
the Pd square plane (sum of six palladium–ligand angles:
699.228, X2 deflection: 14.08) compared to that of complex 4,
suggesting the potential for superior reactivity.
E-mail: stoltz@caltech.edu
[**] We are grateful to the NIH-NIGMS (R01 GM65961-01), NSF
(predoctoral fellowship to D.C.E.), NDSEG (predoctoral fellowship
to D.C.E.), Bristol-Myers Squibb and the American Chemical Society
(predoctoral fellowship to R.M.T.), EPSRC, and EU.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200801865.
Angew. Chem. Int. Ed. 2008, 47, 6367 –6370
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6367

Communications
Table 2: Resolution of a variety of alcohols with [Pd(sp)Br2] (6).
Previously, PdBr2 had been examined as a palladium
source in this transformation, but rapid formation of a black
mixture and low conversion indicated catalyst decomposition
at 808C. Use of preformed complex 6 is only slightly more
successful (Table 1, entry 1). However, in reactions per-
formed at 608C or below, dibromide complex 6 is quite
stable and catalyzes facile oxidation of secondary alcohols
(Table 1, entry 2) relative to dichloride complex 4 (Table 1,
entry 3).[14] Importantly, high selectivity (s) is maintained in
the reactions with dibromide 6.[15] Analogous to our experi-
ments with dichloride complex 4 (Table 1, entry 4),[3] chloro-
form proved to be an excellent solvent, affording product
ketone at good rates at 238C (Table 1, entry 5).[16]
Entry
Alcohol[a]
t [h]
Conv. [%][b]
(yield [%])[c]
Alcohol
s
ee [%][d]
1
4
5
56 (43)96
55
28
95
2[e]
27
20
3
4
8
59 (41)95
59
17
97
4[f]
5[g]
24
60
98
20
13
6
41
30
64 (35)97
63
14
96
Table 1: Optimization of conditions with [Pd(sp)X2].[a]
7[e]
8
24
21
60 (40)93
65
14
99
9[e]
15
12
Entry Solvent T [8C] Pd source
t [h] Conv. Alcohol
[%][b] ee [%][c]
s
10
11
15
48
60
62
91
97
1[d,e]
2
PhCH3
PhCH3
PhCH3
CHCl3
CHCl3
80
60
60
23
23
[Pd(sp)Br2] (6)76
[Pd(sp)Br2] (6)
[Pd(sp)Cl2] (4)24
[Pd(sp)Cl2] (4)48
[Pd(sp)Br2] (6)
32
27
5
9
58
99
29
28
3
54
60
90
99
21
31
4[f]
5[f]
16
15
4
56
96
12
49
45
58 (40)91
58
15
91
[a] Pd source (5 mol%), (À)-sparteine (15 mol%), O2 (1 atm), 0.25m in
solvent, unless otherwise noted. [b] Determined by GC analysis.
[c] Determined by chiral HPLC methods.[10] [d] 0.1m in PhCH3. [e] Pd
black observed. [f] (À)-Sparteine (7 mol%), Cs2CO3 (40 mol%).
3Š M.S.=3 Š molecular sieves; sp=(-)-sparteine.
13[e]
[a] Major enantiomer shown. [b] Determined by GC or 1H NMR methods.
[c] Yield of the isolated enantioenriched alcohol. [d] Determined by chiral
HPLC or GC methods.[10] [e] Performed under ambient air. [f] Performed
at 108C. [g] Performed at 48C.
Oxidative kinetic resolution of a number of secondary
alcohols was facile with this PdBr2 system (Table 2). Alcohols
previously resolved with dichloride 4[3] are oxidized much
more rapidly with dibromide 6 (Table 2, entries 1, 3, and 5),
and the selectivity factors increase at lower temperatures
(Table 2, entries 4 and 5). To our delight, a variety of
secondary alcohols that displayed very poor reactivity with
4 are readily oxidized by using catalyst 6. Sterically hindered
benzylic alcohols (Table 2, entries 6–10), allylic alcohols
(Table 2, entry 11), and even less activated saturated alcohols
(Table 2, entries 12 and 13) are resolved to high enantiomeric
excesses. Furthermore, the use of ambient air instead of pure
oxygen as the stoichiometric oxidant is sufficient for a
successful resolution (Table 2, entries 2, 7, 9, and 13).
led us to diamine 11.[18] Prepared in a three-step sequence
from the easily accessible alkaloid (À)-cytisine, 11 was shown
to act as a (+)-sparteine mimic in a variety of processes.[19,20]
As in the case of palladium complexes with (À)-2, X-ray
crystallographic analyses of [Pd(diamine)X2] reveal a greater
counterion distortion for Br compared to Cl with diamine 11
(Figure 1, sums of six palladium–ligand angles and X1
deflection for 12: 704.678 and 11.98, respectively, and for 13:
701.698 and 14.28, respectively).[13] Indeed, reactions per-
formed with the two catalysts led to disparate results favoring
dibromide 13 (Scheme 2).
Encouraged by our success in promoting rapid oxidation
with a palladium bromide complex, we began to explore other
diamines in the kinetic resolution. Although the availability
of (À)-2 made it an attractive chiral ligand for the process, the
scarcity of its enantiomer (i.e. (+)-2) was a major limitation to
the broad utility of the method.[17] The insights gained from
the investigation of complexes of (À)-2, specifically: A) the
importance of an electron-rich, rigid ligand and B) the need
for an aerobically stable chiral ligand able to induce halide
counterion distortion in its corresponding palladium complex,
Figure 1. X-ray crystallographic structures of [Pd(diamine)X2].
6368 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6367 –6370

Angewandte
Chemie
Table 3: Resolution of alcohols with [Pd(CH3CN)2Br2] and diamine 11.
Entry
Alcohol[a]
t [h]
Conv. [%][b]
Alcohol
s
(Yield [%])[c]
ee [%][d]
1
30
34
58 (40)97
58
25
96
Scheme 2. Kinetic resolution of (Æ)-9 with PdX2 and diamine 11.
2[e]
22
3
30
34
60
61
98
98
19
19
4[e]
Gratifyingly, complex 13 could be generated in situ and
resulted in greatly improved reactivity (Table 3). Thus, a
variety of benzylic (Table 3, entries 1–5), allylic (Table 3,
entries 6–10), and cyclopropylcarbinyl (Table 3, entries 11
and 12) alcohols can be resolved with high selectivity.
Ambient air is also a suitable oxidant (Table 3, entries 2 and
4). Importantly, this protocol yields alcohols in the opposite
enantiomeric series to that obtained with (À)-2.
5
24
61 (38)90
11
6
7
46
12
57 (43)91
55
17
To additionally test these results, we investigated a
synthetically interesting substrate that had proven challeng-
ing for our PdCl2 system. Recently, we reported an enantio-
selective total synthesis of (+)-amurensinine ((+)-3) which
employed an oxidative kinetic resolution as the key step to
produce an enantioenriched intermediate ((À)-14) en route
to the antipode of the natural product.[21] To access the
naturally occurring enantiomer ((À)-3), we applied diamine
11 to the resolution of (Æ )-14. The use of the dibromide
complex provided enantioenriched alcohol (+)-14 in high
yield and excellent enantiomeric excess (s = 27, Scheme 3).
This application constitutes a formal total synthesis of natural
(À)-amurensinine ((À)-3).
94
27
12
8
18
63
94
9
46
35
59 (39)91
63
13
10
92
11
10
In conclusion, we have developed a greatly improved
oxidative kinetic resolution of secondary alcohols based on an
understanding of the factors that contribute to the reactivity
and the selectivity in this process. Furthermore, these
improvements have allowed us to employ alternative diamine
11
12
32
35
59 (40)90
62
13
90
[a] Major enantiomer shown. [b] Determined by GC or 1H NMR methods.
[c] Yield of isolated enantioenriched alcohol. [d] Determined by chiral
HPLC or GC methods.[10] [e] Performed under ambient air.
ligand 11 in the oxidation to afford alcohols in the enantio-
meric series opposite to that obtained with (À)-sparteine.
Importantly, solid-state X-ray crystallographic analysis was
used extensively as a guide in these studies and they provided
invaluable insights. Finally, this methodology was applied to
the kinetic resolution of alcohol (Æ )-14, allowing access to the
natural enantiomer of the isopavine alkaloid (À)-amuren-
sinine. Efforts to additionally enhance reactivity and selec-
tivity in these oxidations, to use this method in complex
molecule synthesis, and to apply these findings to other
palladium-catalyzed oxidations are ongoing.
Received: April 22, 2008
Scheme 3. Preparation of (+)-14. TIPS=triisopropylsilyl.
Published online: July 14, 2008
Angew. Chem. Int. Ed. 2008, 47, 6367 –6370
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6369

Communications
[9] R. J. Nielsen, J. M. Keith, B. M. Stoltz, W. A. Goddard III, J. Am.
Chem. Soc. 2004, 126, 7967 – 7974.
[10] See the Supporting Information for details.
Keywords: alcohols · asymmetric catalysis · oxidation ·
palladium · synthetic methods
.
[11] Oxidations of (+)-9 conducted with [Pd(sp)X2] as the catalyst
displayed the following trend in reactivity: Br> Cl > O2CCF3 >
I > OAc, see the Supporting Information for details.
[12] J. A. Mueller, M. S. Sigman, J. Am. Chem. Soc. 2003, 125, 7005 –
7013.
[13] CCDC 298214 for 6, 274539 for 12, and 639648 for 13 contain
the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[14] Poor reactivity and selectivity were observed in the oxidative
kinetic resolution with [Pd(sp)I2] as the catalyst, see the
Supporting Information for details.
[15] The selectivity factor s was determined by using the equation s =
kfast/kslow = ln[(1ÀC)(1Àee)]/ln[(1ÀC)(1+ee)], where C is con-
version, see: reference [6a].
[16] The reactivity of 6 in chloroform was comparable to complexes
prepared in situ from a variety of PdBr2 sources, see the
Supporting Information for details.
[17] For an asymmetric synthesis of (+)-sparteine, see: B. T. Smith,
J. A. Wendt, J. Aubꢀ, Org. Lett. 2002, 4, 2577 – 2579.
[18] a) M. J. Dearden, C. R. Firkin, J.-P. R. Hermet, P. OꢁBrien, J.
Am. Chem. Soc. 2002, 124, 11870 – 11871; b) M. J. Dearden, M. J.
McGrath, P. OꢁBrien, J. Org. Chem. 2004, 69, 5789 – 5792.
[19] A. J. Dixon, M. J. McGrath, P. OꢁBrien, Org. Synth. 2006, 83,
141 – 154.
[1] a) B. M. Stoltz, D. C. Ebner in Handbook of C-H Transforma-
tion, Vol. 2 (Ed.: G. Dyker), Wiley-VCH, New York, 2005,
pp. 393 – 401; b) B. M. Stoltz, Chem. Lett. 2004, 33, 362 – 367;
c) M. S. Sigman, D. R. Jensen, Acc. Chem. Res. 2006, 39, 221 –
229.
[2] a) E. M. Ferreira, B. M. Stoltz, J. Am. Chem. Soc. 2001, 123,
7725 – 7726; b) J. T. Bagdanoff, E. M. Ferreira, B. M. Stoltz, Org.
Lett. 2003, 5, 835 – 837.
[3] J. T. Bagdanoff, B. M. Stoltz, Angew. Chem. 2004, 116, 357 – 361;
Angew. Chem. Int. Ed. 2004, 43, 353 – 357.
[4] Simultaneous with our report, a similar system was reported,
see: D. R. Jensen, J. S. Pugsley, M. S. Sigman, J. Am. Chem. Soc.
2001, 123, 7475 – 7476.
[5] For some synthetic applications, see: a) D. D. Caspi, D. C. Ebner,
J. T. Bagdanoff, B. M. Stoltz, Adv. Synth. Catal. 2004, 346, 185 –
189; b) D. C. Ebner, Z. Novµk, B. M. Stoltz, Synlett 2006, 3533 –
3539.
[6] For discussions of kinetic resolution, see: a) H. B. Kagan, J. C.
Fiaud in Topics in Stereochemistry, Vol. 18 (Ed.: E. L. Eliel),
Wiley, New York, 1988, pp. 249 – 330; b) J. M. Keith, J. F. Larrow,
E. N. Jacobsen, Adv. Synth. Catal. 2001, 343, 5 – 26; c) E. Vedejs,
M. Jure, Angew. Chem. 2005, 117, 4040 – 4069; Angew. Chem.
Int. Ed. 2005, 44, 3974 – 4001.
[7] For examples with alternative ligands in the palladium-catalyzed
oxidative kinetic resolution of secondary alcohols, see: a) D. R.
Jensen, M. S. Sigman, Org. Lett. 2003, 5, 63 – 65; b) T. Chen, J.-J.
Jiang, Q. Xu, M. Shi, Org. Lett. 2007, 9, 865 – 868.
[8] R. M. Trend, B. M. Stoltz, J. Am. Chem. Soc. 2004, 126, 4482 –
4483.
[20] P. OꢁBrien, Chem. Commun. 2008, 655 – 667.
[21] U. K. Tambar, D. C. Ebner, B. M. Stoltz, J. Am. Chem. Soc. 2006,
128, 11752 – 11753.
6370 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6367 –6370